A long-term daily 500 m snow cover extent product over China (1981–2000)

Due to the limitations of early satellite sensors in terms of spatial resolution, temporal resolution, and time series availability, there is currently a lack of long-term daily snow cover extent (SCE) products with 500 m spatial resolution prior to 2000. To address this data gap, this study extends the existing SCPOT algorithm across China and incorporates methodological improvements. First, the similar conditional probability (SCP) between Moderate Resolution Imaging Spectroradiometer (MODIS) SCE and Advanced Very High Resolution Radiometer (AVHRR) SCE was calculated across different spatial scales during overlapping periods. Based on the AVHRR SCE from 1981 to 2000, a series of downscaled daily SCE products at 500 m resolution were generated. Validation against ground-based snow depth observations yielded an overall accuracy (OA) of 0.84, a recall (RC) of 0.77, a precision (PC) of 0.93, and a Cohen’s kappa (CK) of 0.69. Further validation with contemporaneous Landsat-5 SCE showed OA, RC, PC, and CK values of 0.82, 0.83, 0.81, and 0.63, respectively. Accuracy assessments reveal that the product maintains high inter-annual stability throughout the time series, with OA consistently above 0.8 and CK around 0.7. These results collectively demonstrate the reliability and stability of the downscaled SCE product, effectively filling this historical data gap.